Geography Reference
In-Depth Information
tive to behave opportunistically because of loss of reputation regarding the common
third.
Note here that the role of social proximity in the formation of network links relates
to the concept of closure. Viewing social proximity between two actors as the inverse
of geodesic distance (network distance) in a network, closure simply means that if two
actors have a social distance of two, they have a higher probability of getting connected.
More generally, one expects the probability of a link to be formed to decrease with an
increase in geodesic distance between two actors. Dynamically, this means that one
expects the social proximity in networks to increase over time.
One hypothesis that has been analysed in a study by Ter Wal (2009) holds that closure
in particularly relevant as a mechanism of network formation in exploitation contexts,
while it is less important in exploration contexts. The reasoning underlying this hypoth-
esis holds that closure is a way to i nd a new partner through an existing trusted partner,
so that the collaboration with the new partner is embedded in the common relationship
with the third actor. As a result, the partners in the new collaboration will have less
incentive to behave opportunistically as they risk jeopardizing their relation with the
third actor. Avoiding opportunistic behaviour is especially important in the exploitation
phase of an industry, during which knowledge becomes more codii ed and is transformed
in commercial products and services and, consequently, trust in partners is most impor-
tant. And, logically, the formation of a new network relation is more likely, the more two
actors already have partners in common. Studying the evolution of co-inventor networks
in the German biotech industry in the period 1970 to 1995, Ter Wal (2009) found that
closure was indeed a key factor driving network formation. As expected, closure also
became more important over time as a driver of network formation with the biotech
sector evolving from the exploration to the exploitation phase.
Proximity, industrial lifecycle and regional lock-in
As explained, the dif erent forms of proximity likely inl uence the decisions of agents
with whom to connect. As individuals and organizations prefer to establish relationships
with similar type of individuals or organizations, network clustering will result as similar
actors group together. When linking the proximity concept to the geography of innova-
tion networks, major research challenges remain to be taken up. These have not yet (or
hardly) been explored in the network literature, but are essential for the development of
an evolutionary approach to the geography of innovation networks.
The main challenge is the study of the dynamics of network formation: how innovation
networks of i rms evolve in time and space, and what forms of proximity are important
at what stage of the evolution of the network. The focus of attention is on the dynamics
in the number of nodes and relations, and how the dif erent forms of proximity impact
on these network dynamics. It concerns both the study of: (1) the creation of relations by
new i rms entering the industry and by incumbent i rms linking up with other nodes; and
(2) the break-up of existing relations because of the exit of i rms or because incumbent
i rms dissolve their relations with other nodes. Doing so, it covers the process of crea-
tive destruction proposed by Schumpeter, and applies that to the evolution of networks.
Such an approach also accounts for the evolutionary concept of selection that basically
takes place at two levels: (1) the impact of competition on i rms leading to i rm dynam-
ics (i.e. the entry and exit of nodes); (2) the choice of linking or breaking with network
